Increased numbers of activated brain MP (perivascular and brain macrophages and resident microglia) remain the principal histologic correlate for dementia in late-stage HIV- disease. We hypothesize that in HAD HIV-1-infected and immune-activated MP secrete soluble toxins that damage neural tissue, ultimately causing cognitive and motor dysfunction. Two aspects of this hypothesis remain poorly understood. First, mechanisms underlying brain MP activation in HAD are not known. Recently, we demonstrated increased expression of the immune molecule CD40 ligand (CD40L) on peripheral blood mononuclear cells from late-stage HIV-1 patients. Further, we showed that CD40L activates primary human macrophages in vitro, an effect that is potentiated by HIV-1 infection. Second, mechanisms linking MP activation to neuronal dysfunction are not well defined. Two important classes of molecules, chemokines and excitatory neurotoxins, are secreted by HIV-1-infected, immune-activated macrophages and may alter neuronal function through receptor-mediated or non-specific mechanisms. Indeed, human neurons express receptors for both chemokines and glutamate. In toto,we believe that CD40Lstimulation of virus-infected brain MP induces the production of chemokines and glutamate, which ultimately lead to neuronal compromise. Using primary human macrophage and neuronal culture systems, we propose to study mechanisms that underlie macrophage activation in HAD and link macrophage activation to neuronal dysfunction. This proposal integrates disciplines of virology, immunology, and receptor pharmacology in a mechanistic study of HAD.